Retroviremia in commercial pigs and its preliminary association with poor health.

We report the novel discovery of age-related titers of circulating retroviral elements in pig serum and a preliminary association with background disease status. Exogenous retroviruses have previously not been described for pigs; however, porcine endogenous retroviruses (PERV) are well characterized due to their potential significance and the risks associated with microbiological safety in clinical xenotransplantation. Three subtypes of PERV have been identified in the pig genome. PERV A and B are present in the genomes of all pigs, and immortalized pig cell lines were able to produce virus capable of infection of certain pig and human cell lines (8). In contrast, PERV C is absent from many pigs and this virus is not infectious to human cells. More recently, the possible existence of exogenous (and therefore transmissible) as opposed to endogenous retroviruses (ERV) in pigs was proposed by Scobie et al. (6, 9). They described a PERV A/PERV C recombinant retrovirus released from mitogen-stimulated porcine peripheral blood mononuclear cells that was not present in a proviral form in the pig genome. ERVs account for around 5% of the vertebrate genome (3) and represent previously exogenous retroviruses that have integrated into the host germ line with evolutionary, physiological, and pathological consequences (5). Human ERV have been associated with neoplasia, congenital defects, and autoimmune diseases (4), and retrovirus-like particles have been described for human fetal serum (2) and synovial fluid from cases of rheumatoid arthritis (7). To date, few studies have investigated a causal relationship between ERV and disease and there is no evidence for retrovirus-associated pathogenesis in pigs. Further, ERV expression has been considered a marker of disease, through secondary activation, rather than a causal agent (4). However, in mice, an immune complex-mediated nephritis has been causally associated with ERV (1). Therefore, in pigs, retroviremia (endogenous, exogenous, or both) might be associated with increased susceptibility to disease, either through directly pathogenic effects or through immune dysfunction. The current study investigated the prevalence and quantity of viral RNA in serum as an indicator of retroviremia in commercial pigs. The study used a real-time reverse transcriptase PCR (RT-PCR) approach, with primers (kindly supplied by Prof. D. Onions, Q-One Biotech Ltd., Glasgow, Scotland) directed at a pol gene sequence conserved across all currently known PERV A, B, and C types, to analyze viral RNA extracted from 87 pig sera. The sera originated from four farms: farm 1, a specific pathogen-free colony of Danish Ellegaard minipigs (free from porcine reproductive and respiratory syndrome virus [PRRSV] and postweaning multisystemic wasting syndrome [PMWS]); farm 2, a United Kingdom Large White cross herd (free from PRRSV and PMWS); farm 3, a United Kingdom Large White cross herd, positive for PRRSV and PMWS; and farm 4, a Spanish Large White cross herd, negative for PRRSV and PMWS. The results (Fig. ​(Fig.1a)1a) show that serum-associated PERV RNA was detected in all but one (86/87) of the serum samples (range, 0 to 5.74 × 106 copies/ml serum). A univariate analysis of variance (ANOVA; SPSS 13.0), performed with PERV RNA serum titer (as a response variable) measured on the log scale in order to meet ANOVA assumptions, and farm and age as response variables indicated significant differences in PERV RNA serum titer in consideration of farm (P < 0.0001) and age of pig sampled (P < 0.0001). There was also strong evidence of an interaction between farm and age (P = 0.057), although this is not quite significant at the 5% level. FIG. 1. (a) Box whisker plots of PERV RNA detection in pig serum samples (RNA copies/ml serum) collected from pigs of defined age groups on four different farms (a total of 87 pigs across four farms). On farm 1 (Danish minipigs, high health), 6 pigs >6 ... Overall, there was a tendency for increased levels of serum viral RNA in pigs aged 20% for farm 3, ∼10% for farm 4, and <5% for farms 1 and 2 (M. Donadeu, personal communication). Preliminary characterization of virus present was carried out by amplification of serum RNA with primers specific for a conserved region of PERV env (6). False-positive results arising from contaminating cellular, proviral DNA were excluded by omitting the RT enzyme (data not shown). Preliminary sequencing analysis of the RT-PCR products revealed homology with an env gene sequence from a previously described PERV A/C recombinant retrovirus (9) (Fig. ​(Fig.1b).1b). The findings are unlikely to represent leaked, cellular mRNA as the extraction method-targeted viral as opposed to mammalian RNA, the latter of which is considered relatively unstable in serum and likely to be degraded before testing. These data support the existence of retroviremia, which includes the A/C recombinant. This variant has been proposed to be exogenous in miniature swine (6, 9). Extended analysis may reveal further variants, including PERV, but we cannot exclude the possibility of the presence of a yet-uncharacterized exogenous retrovirus in pigs. Additional studies are required to confirm and better characterize these viruses in relation to a causal direct or indirect pathogenicity associated with PMWS and PRRSV or other endemic diseases, irrespective of whether they have exogenous or endogenous origins. These analyses should also assess the significance of background genetics, epidemiology of transmissible agents, and the implications for human public health as well as pig production.